Special Issues
Table of Content

Privacy-Preserving Deep Learning and its Advanced Applications

Submission Deadline: 31 December 2024 View: 644 Submit to Special Issue

Guest Editors

Prof. Celimuge Wu, The University of Electro-Communication, Japan
Prof. Soufiene Djahel, Coventry University, UK
Prof. Jie Feng, Xidian University, China
Prof. Kok-Lim Alvin Yau, Universiti Tunku Abdul Rahman, Malaysia
Prof. Yange Chen, Xuchang University, China

Summary

Deep learning has demonstrated remarkable accomplishments in various domains, including speech recognition and image processing. While deep learning offers novel solutions for these applications, its training model necessitates a substantial volume of data. Consequently, service providers are compelled to amass extensive participant data, which may encompass confidential details pertaining to companies or users, such as medical records, account information, and business operations. The utilization of this sensitive information in deep learning applications can potentially result in the inadvertent exposure of sensitive data of enterprises or users. Simultaneously, as users become more cognizant of safeguarding their personal privacy and governments implement stricter information security laws and regulations, there is a growing emphasis in various domains on investigating multi-participant privacy-preserving deep learning.


The objective of this topical collection is to solicit novel contributions pertaining to recent advancements in privacy-preserving deep learning. This special issue endeavors to encompass a broad spectrum of subjects associated with the practical implementation of deep learning, encompassing, but not limited to:

· Privacy-preserving deep learning via homomorphic encryptions

· Privacy-preserving deep learning via differential privacy

· Privacy-preserving deep learning via secure multi-party computation

· Applications of privacy-preserving deep learning

· Privacy-preserving distributed deep learning

· Privacy-preserving federated deep learning and its applications

· Privacy-preserving deep reinforcement learning

· Deep unlearning and its applications

· Privacy-preserving deep learning via other privacy technologies

· Privacy-preserving active learning for training deep learning

· Distributed computing for privacy-preserving deep learning

· Performance optimization of privacy-preserving deep learning

· Experimental testbeds of privacy-preserving deep learning 


Keywords

Deep learning; Privacy-preserving; Distributed deep learning; Federated deep learning; Deep reinforcement learning; Deep unlearning; Performance optimization

Published Papers


  • Open Access

    ARTICLE

    Privacy Preservation in IoT Devices by Detecting Obfuscated Malware Using Wide Residual Network

    Deema Alsekait, Mohammed Zakariah, Syed Umar Amin, Zafar Iqbal Khan, Jehad Saad Alqurni
    CMC-Computers, Materials & Continua, Vol.81, No.2, pp. 2395-2436, 2024, DOI:10.32604/cmc.2024.055469
    (This article belongs to the Special Issue: Privacy-Preserving Deep Learning and its Advanced Applications)
    Abstract The widespread adoption of Internet of Things (IoT) devices has resulted in notable progress in different fields, improving operational effectiveness while also raising concerns about privacy due to their vulnerability to virus attacks. Further, the study suggests using an advanced approach that utilizes machine learning, specifically the Wide Residual Network (WRN), to identify hidden malware in IoT systems. The research intends to improve privacy protection by accurately identifying malicious software that undermines the security of IoT devices, using the MalMemAnalysis dataset. Moreover, thorough experimentation provides evidence for the effectiveness of the WRN-based strategy, resulting in… More >

  • Open Access

    ARTICLE

    Two-Stage Client Selection Scheme for Blockchain-Enabled Federated Learning in IoT

    Xiaojun Jin, Chao Ma, Song Luo, Pengyi Zeng, Yifei Wei
    CMC-Computers, Materials & Continua, Vol.81, No.2, pp. 2317-2336, 2024, DOI:10.32604/cmc.2024.055344
    (This article belongs to the Special Issue: Privacy-Preserving Deep Learning and its Advanced Applications)
    Abstract Federated learning enables data owners in the Internet of Things (IoT) to collaborate in training models without sharing private data, creating new business opportunities for building a data market. However, in practical operation, there are still some problems with federated learning applications. Blockchain has the characteristics of decentralization, distribution, and security. The blockchain-enabled federated learning further improve the security and performance of model training, while also expanding the application scope of federated learning. Blockchain has natural financial attributes that help establish a federated learning data market. However, the data of federated learning tasks may be… More >

Share Link